<< 25-04-2017 >>
00:04:45*Sentreen joined #nim
00:11:07FromGitter<nortiero> Hi! I am playing with Nim types... Given `type RefInt = ref int` it is possible to allocate *and* initialize a new RefInt ... something like var v = RefInt(42) ?
00:14:46FromGitter<Varriount> nortiero: I don't believe it can be done with a direct reference to a basic type like int
00:15:23FromGitter<Varriount> If you had a reference to an object, you could do `var m = RefFoo(memberOne: 1)`
00:15:50FromGitter<nortiero> yes, that's what i've been doing, but here .. the field is anonymous
00:16:16FromGitter<nortiero> i have to call new and then set with x[] = 42
00:16:28FromGitter<Varriount> You could create a procedure.
00:16:50FromGitter<Varriount> It will likely be inlined by the compiler. Or you could add {.inline.}
00:17:28FromGitter<nortiero> of course! I was just playing with types. This kind of ref type reminds me of things seen in Rust, and I was just curious... Thank you!
00:18:43FromGitter<Varriount> Although, creating a reference to an integer isn't common. A sequence of integers or a reference to an array is much more common.
00:19:21FromGitter<nortiero> It has little use and a lot of overhead
00:43:19*krux02 joined #nim
00:53:55FromGitter<hcorion> Hi all. I'm using a rather large codegenDecl pragma multiple times in my code, would it be possible to declare a custom pragma, that basically functions as that pragma?
00:56:09FromGitter<Varriount> @hcorion Say hello to the somewhat odd "pragma pragma": https://nim-lang.org/docs/manual.html#pragmas-pragma-pragma
00:56:52FromGitter<hcorion> Awesome thanks!
00:58:36zachcarterman…
00:58:50zachcarterrunning my program as a service has to be easier than what I’m doing right now
00:58:50FromGitter<Varriount> zachcarter: ?
00:58:59zachcarteron digital ocean
00:59:46zachcarterI’m trying to do what’s suggested here : https://github.com/nim-lang/Nim/wiki/Tutorial:-Creating-a-(micro)-service
01:00:10FromGitter<hcorion> @Varriount Do you know how I can set the "pragma pragma" as public?
01:00:13zachcarterbut the service just keeps crashing I’m guessing because I don’t have permissions correct
01:00:44FromGitter<Varriount> @hcorion You'll need to use a macro for that, I think.
01:01:11FromGitter<Varriount> hcorion: Do you want the quick and dirty solution, or something elegant?
01:01:42FromGitter<hcorion> elegant would be nice
01:02:54FromGitter<Varriount> Zion: If this is for procedures only, have a macro that is used as a pragma (https://nim-lang.org/docs/manual.html#macros-macros-as-pragmas). Have the macro modify the procedure's AST so that the desired pragmas are injected.
01:03:34FromGitter<Varriount> If this is a statement macro, I think you have to use a template to produce the entire pragma.
01:04:16FromGitter<hcorion> Yeah, it's variables, not procs
01:05:23FromGitter<Varriount> Zion: Then just create a template with a body that produces the macro
01:05:59FromGitter<Varriount> I don't know if you can use macros in statement pragma lists
01:06:14FromGitter<Varriount> Zion: Could I get an example?
01:06:55FromGitter<hcorion> https://gist.github.com/hcorion/be452250aeb46e02b1dd4ad5ae519d8d
01:24:44FromGitter<Varriount> @hcorion https://glot.io/snippets/ep952xc9xx
01:26:04FromGitter<Varriount> @hcorion The alternative is to create a single module with that pragma, and use `include` wherever you want the pragma declared.
01:26:36FromGitter<Varriount> You might want to see if a bug/feature request has been lodged for macros to work in statement/variable pragma lists
01:32:04*bjz quit (Quit: My MacBook Pro has gone to sleep. ZZZzzz…)
01:32:32*chemist69 quit (Disconnected by services)
01:32:37*chemist69_ joined #nim
01:34:18*htnjns joined #nim
01:42:19*ludocode quit (Ping timeout: 255 seconds)
01:50:10*vlad1777d quit (Quit: Leaving)
01:53:05*ludocode joined #nim
01:56:06*mwbrown quit (Quit: Exiting)
02:35:19*bjz joined #nim
02:44:28*sz0 quit (Quit: Connection closed for inactivity)
02:46:57*mwbrown joined #nim
02:47:07*brson quit (Quit: leaving)
02:55:42*krux02 quit (Remote host closed the connection)
02:56:25*yglukhov joined #nim
03:00:32*yglukhov quit (Ping timeout: 240 seconds)
03:02:23*mwbrown quit (Quit: Exiting)
03:15:15*mwbrown joined #nim
03:16:18*bjz quit (Quit: My MacBook Pro has gone to sleep. ZZZzzz…)
03:25:52*bjz joined #nim
03:32:13*zachcarter quit (Quit: zachcarter)
03:46:19*mwbrown quit (Quit: Exiting)
03:57:30*mwbrown joined #nim
04:02:29*mwbrown quit (Ping timeout: 260 seconds)
04:06:28shashlickhow do you pass an array by reference?
04:08:00*mwbrown joined #nim
04:18:36*mwbrown quit (Ping timeout: 260 seconds)
04:33:00*mwbrown joined #nim
04:37:51*mwbrown quit (Ping timeout: 260 seconds)
04:41:29*mwbrown joined #nim
04:50:05*BennyElg quit (Remote host closed the connection)
04:54:48*bjz quit (Quit: My MacBook Pro has gone to sleep. ZZZzzz…)
05:00:48*bjz joined #nim
05:16:35*gokr joined #nim
05:44:01FromGitter<Varriount> shashlick: Why?
05:44:47FromGitter<Varriount> If you're concerned about optimization, Nim already automatically passes data types above a certain size by pointer.
05:54:51*rokups joined #nim
05:55:32*Snircle quit (Quit: Textual IRC Client: www.textualapp.com)
06:26:12*Vladar joined #nim
06:44:42*bjz quit (Quit: My MacBook Pro has gone to sleep. ZZZzzz…)
06:47:36*bjz joined #nim
06:50:53*nsf joined #nim
06:59:33*yglukhov joined #nim
07:00:53*Arrrr joined #nim
07:00:53*Arrrr quit (Changing host)
07:00:53*Arrrr joined #nim
07:08:05*sz0 joined #nim
07:19:04*BennyElg joined #nim
07:23:25*BennyElg quit (Ping timeout: 255 seconds)
07:28:16*BennyElg joined #nim
07:30:32*Sembei quit (Ping timeout: 240 seconds)
07:37:36*Sembei joined #nim
07:38:07*vendethiel- joined #nim
07:39:34*chemist69_ quit (Quit: WeeChat 1.7)
07:40:48*antranig1 joined #nim
07:41:11*chemist69 joined #nim
07:41:32*libman quit (Quit: Connection closed for inactivity)
07:41:32*antranigv quit (Remote host closed the connection)
07:42:31*Sembei quit (Ping timeout: 240 seconds)
07:44:13chemist69shashlick: you can pass the array as var parameter: `proc myproc(myarr: var MyArray) = ...`
07:44:31chemist69then it is modifiable by the called function
07:46:16*vendethiel quit (Ping timeout: 260 seconds)
08:01:44*Jesin quit (Ping timeout: 258 seconds)
08:02:02*antranig1 is now known as antranigv
08:18:21*nightmared quit (Ping timeout: 260 seconds)
08:19:59*couven92 joined #nim
08:44:57ArrrrWhy is not possible to index arrays using ordinals? I guess this is a bug, an old one https://glot.io/snippets/ep9h79x3b1
08:45:32ArrrrSame result with let
08:48:28Araqtry to annotate your const with an explicit array type
08:50:31ArrrrYeah it works https://glot.io/snippets/ep9hczlxyd but it is more work than simple `NUMS[ord(dir)]`
08:55:52FromGitter<stisa> Arrrr doesn't ``echo NUMS[ord UP]`` work? Or am I misunderstanding what you want to do?
08:57:59*Vladar quit (Quit: Leaving)
09:00:49ArrrrYes, but it should work without ord. From the snippet, nim should implicit understand the const is of type array[enum, int], not array[int, int]. I don't have any problems with adding ord, but it looks unnecessary to me.
09:07:53AraqArrrr: yeah, the type inference is wrong here
09:11:36*Vladar joined #nim
09:13:23*McSoFake joined #nim
09:23:48*zachcarter joined #nim
09:25:06zachcarterugh up way too early
09:25:47couven92zachcarter, really? it's already 11:25 at my end! :P
09:25:58zachcarteroh man!
09:26:02zachcarteronly 5:25 am here!
09:26:23couven92That IS early! My condolences!
09:26:46zachcarterI blame the dogs!
09:26:58zachcarterbut thank you
09:27:18zachcarterI guess now I have more time to work on this service / digital ocean crap I’ve been struggling to figure out
09:27:37zachcarterdom96: I found out last night jsonp doesn’t support POST requests so it’s a nogo
09:27:57Araqwhat is 'jsonp'?
09:28:17couven92well I am successfully putting of my upcoming battle with JavaScript... Ugh! :P
09:28:27flyxJSON which is fetched as body of a <script src=""> tag
09:28:48flyxone of the ugliest things web devs have ever done
09:29:03zachcarteravoids x domain requests though
09:29:09zachcarterwhich is why I guess dom was suggesting it
09:29:12flyxyes
09:29:22*Tiberium joined #nim
09:29:27couven92flyx, how does that work? Do you assign an id to the tag and then through JavaScript set it's contents?
09:29:28zachcarterI’ll have to just fight with CORS
09:29:41zachcarteryou createa script tag dynamically
09:29:47flyxit doesn't avoid x domain requests, it's just that the standard says that x domain requests from <script> tags are allowed
09:29:48zachcarterat runtime
09:29:58zachcarteryeah what flyx said :P
09:30:58couven92dynamically add script tags... Isn't that what we call self generating code? :O
09:31:34zachcarterI mean I guess you could alter the contents of an existing scirpt tag
09:31:36zachcarterscript tag
09:31:37zachcarterand execute it
09:32:02couven92like so many things in JS, this does not sound like a good idea! :P
09:32:07zachcarterif you google jsonp javascript you’ll find examples of people doing it
09:32:15zachcarterJS is a horror show
09:34:12euantorWhy do you need cross-domain AJAX requests anyway?
09:34:29euantorWhenever I've had that issue I've found better ways around it
09:35:13zachcarterin case someone wants to hit the API from another domain
09:36:02euantorHm. Yes, that could be a problem
09:36:14zachcarteryeah :/
09:36:22zachcarterthis is for the nim playground thing euantor
09:36:41euantorYeah, I guessed that. Does it have to be a POST request?
09:37:13zachcarterI think it makes sense for it to be one
09:37:45zachcarterI could do ugly query parameter stufff with a GET request but...
09:37:47euantorYeah, I guess CORS is the only option then realistically
09:38:22euantorEasiest hackiest approach is `Access-Control-Allow-Origin: *` and `Access-Control-Allow-Methods: POST`
09:38:37zachcarterah I was missing the latter part of that, thank you
09:38:38*GustavoLapasta joined #nim
09:39:05euantorI think there are some other headers about max age and stuff too, but I can't remember what they do
09:39:27zachcarterthis should get me going
09:39:40zachcarterright now I’m fighting getting the service to run as a daemon without crashing instantly upon receiving a post request
09:39:51euantorUsing systemd?
09:39:53zachcarteryup
09:40:01euantorWHat's the crash log look like?
09:40:19zachcarterhttps://gist.github.com/zacharycarter/0f53ea961ea211e485abef3f20da813e
09:40:24zachcarterI’m guessing it’s due to file system isssues
09:40:32zachcarterbecause the process creates a tmp directory
09:40:44zachcarterreads from another directory and writes to the tmp directory
09:40:53euantorPossibly
09:41:15euantorShould be fun to track down
09:41:30zachcarteryeah :/ I’ve never really messed with systemd before so this is painful for me
09:41:47zachcarterI wish services on nix were as easy as they were on windows
09:42:00zachcarteror rather I wish I understood as much about them as I do about services on windows
09:42:19euantorWhat have you got set up as the After and Requires?
09:42:32zachcarterAfter=network.target httpd.service squid.service nfs-server.service mysqld.service named.service postfix.service
09:42:33zachcarterWants=network-online.target
09:43:01euantorthat looks right to me. /tmp/ should be alive by that point I would think
09:43:18zachcarterit’s actually tring to write to ./tmp
09:43:33zachcarterso wherever the process is running from
09:43:47euantorThe stack trace points to the line `await sleepAsync(500)`
09:43:51euantorAh
09:44:01zachcarteryeah I figured it’s because the async method is failing
09:44:23euantorI wonder if you need to set the working directory
09:44:31zachcarterI tried that too :/
09:44:35euantorBecause by default the current working directory will probably be /
09:45:15zachcarterof course everything works fine if I run the service locally
09:45:17zachcarteroh I know what to try!
09:45:25zachcartersorry I meant the binary
09:45:30zachcarterrun the binary as that user let me give that a shot
09:45:53euantorYeah, that's worth a shot
09:49:39FromGitter<Varriount> zachcarter: Do we want people to hot this from another API? Doesn't seem like something that should be done for an interactive playground.
09:50:11zachcarterhrm I think dom96 was mentioning he wanted to potentially use this for other things
09:50:16euantorIt might be handy to have a "Run this code" option in the docs
09:50:19zachcarterI guess he could set up and deploy another instance of the project for that
09:50:21euantorLike Go has
09:50:45zachcarterwe could probably just get pretty restrictive with what domains we allow post requests from
09:50:54zachcarteror if everything is on nim-lang.org
09:50:57zachcarterwe don’t have to worry about it
09:51:09euantorSomething like https://tour.dlang.org/ would be cool too
09:51:20euantor(see the "Run" button in the top right)
09:51:28zachcarterthat would be neat
09:51:38euantorYeah, you could restrict to *.nim-lang.org
09:52:19zachcarterthat could work
09:52:56euantorIIRC you can comma separate domains so `*.nim-lang.org,nim-lang.org` would cover everything probably
09:53:09euantorAnd it would be super cool to be able to run code inside forum threads too
09:53:18zachcarteryeah it would!
09:53:20zachcarteror from irc
09:53:28zachcarterand gitter
09:53:46FromGitter<Varriount> If we had the money, I would use AWS autoscaling and/or lambda
09:53:52euantorYeah, so plenty of use cases
09:54:20euantorI think DigitalOcean has donated some funds to run servers which is what hosts the forum and site at the minute
09:54:30FromGitter<Varriount> Or better yet, have the compiler compile and run the code virtually, within the user's browser
09:54:32*chemist69 quit (Ping timeout: 255 seconds)
09:55:08TiberiumVarriount: nim-sandbox does that
09:55:29Tiberiumhttps://github.com/jscheid/nim-sandbox it works with latest nim devel, but it's configured without any modules except "system"
09:55:30FromGitter<Varriount> euantor: Yes, but such charity costs then little. It's not like Nim's website is a highly dynamic webapp
09:55:52Tiberiumbasically it's compiler compiled via emscripten
09:55:59Tiberiumand then compiler emits JS code in browser
09:56:08euantorTrue, but it should be enough to run the playground I would think
09:56:25*PMunch joined #nim
09:57:55FromGitter<Varriount> For the forum and documentation, caching might help.
09:58:22euantoryeah, would be a good idea
09:58:26FromGitter<Varriount> Since compiler input, as a whole, produces the same output after repeated rubs
09:58:31FromGitter<Varriount> *runs
09:58:55*chemist69 joined #nim
09:59:02euantorcaching would have to be inside the playground code, but shouldn't be overly difficult
09:59:17zachcarteryeah I can add caching
09:59:24Tiberiumyeah, because I think 90% of playground requests would be the same from "samples"
09:59:32Tiberium(I mean there should be samples in nim playground)
09:59:50*adeohluwa joined #nim
10:01:48euantorzachcarter: How well does the playground handle concurrent requests? Looking at it, wouldn't `./tmp/in.nim` get overwritten by a second request if the first request was running?
10:02:13zachcarteryeah I’m going to have to create GUIDs for each request
10:02:46zachcarterI’m working on some modifications to it right now
10:03:00euantorI wonder if it might be a nice addition to the stdlib to add something like https://github.com/rust-lang-nursery/tempdir
10:04:06euantorwith a template like `withTempDir(prefix="nim_playground")`
10:04:30euantorI might have a go at writing that actually
10:04:42Tiberiumseems like very professional benchmarks: https://github.com/rust-lang-nursery/rustc-benchmarks/blob/master/compare.py
10:04:47Tiberiumusing python's time.time()
10:04:55Tiberiumeven python's own benchmarks doesn't use it
10:05:37euantor¯\_(ツ)_/¯
10:09:35FromGitter<Varriount> I might try something like this with lambda
10:12:28*vlad1777d joined #nim
10:17:34euantorHas anybody else found that libraries.io is supporting Nimble now: https://libraries.io/nimble/sysrandom
10:18:34*GustavoLapasta quit (Ping timeout: 268 seconds)
10:19:12Tiberiumeuantor, I think it was doing it a long time ago
10:19:26euantorI'd never heard of it before, it's pretty cool
10:29:25*CcxWrk quit (Ping timeout: 260 seconds)
10:33:42*CcxWrk joined #nim
10:53:24*nightmared joined #nim
11:02:42*Snircle joined #nim
11:04:25*Tiberium quit (Remote host closed the connection)
11:04:46*Tiberium joined #nim
11:05:08Araqhow can I get emscripten to work on OSX?
11:05:59AraqI have python 2.7.7 and yet it fails with
11:06:03Araq File "/usr/local/Cellar/emscripten/1.37.1/libexec/tools/toolchain_profiler.py", line 3
11:06:03Araq EM_PROFILE_TOOLCHAIN = int(os.getenv('EM_PROFILE_TOOLCHAIN')) if os.getenv('EM_PROFILE_TOOLCHAIN') != None else 0
11:06:03Araq ^
11:06:05AraqSyntaxError: invalid syntax
11:06:16*Tiberium quit (Remote host closed the connection)
11:08:13*Tiberium joined #nim
11:09:49*Tiberium quit (Remote host closed the connection)
11:12:11*Tiberium_ joined #nim
11:12:11*Tiberium joined #nim
11:12:14*Tiberium__ joined #nim
11:12:47FromGitter<Varriount> Araq: How are you invoking emscripten?
11:12:47*Tiberium quit (Remote host closed the connection)
11:12:47*Tiberium_ quit (Remote host closed the connection)
11:12:48*Tiberium__ quit (Remote host closed the connection)
11:13:56Araq cc = clang
11:13:56Araq clang.exe = "emcc"
11:13:56Araq clang.linkerexe = "emcc"
11:13:57Araq clang.options.linker = ""
11:13:59Araq cpu = "i386"
11:14:02Araq passC = "-s WASM=1 -s 'BINARYEN_METHOD=\"native-wasm\"' -Iemscripten"
11:14:03Araq passL = "-s WASM=1 -Lemscripten -s TOTAL_MEMORY=335544320"
11:14:24Araqbut regardless of how I invoke it, it shouldn't fail with a Python syntax error...
11:15:11euantorHm. When I try to build documentation for some code using `getTempDir` from the `os` module I get `ambiguous call; both ospaths.getTempDir() and os.getTempDir() match for: ()`
11:15:23euantorPrefixing `getTempDir` as `os.getTempDir()` doesn't help
11:16:30*Tiberium joined #nim
11:17:16Araqeuantor: don't import os and ospaths
11:17:22euantorI don't
11:17:54euantorhttps://www.irccloud.com/pastebin/zQe26kIO/
11:18:00euantorThat's all of my imports
11:18:27euantorhttps://www.irccloud.com/pastebin/xw1VpZOv/
11:18:45euantorI'd assume it's because I'm running it as a nimble task
11:18:59FromGitter<stisa> Araq I think you need to run a bat file that sets up some env variables, I don't remember the name though, `emsdk<something>` I think?
11:19:06euantorNope, happens if I run the command on its own
11:19:31FromGitter<Varriount> Araq: I get "zipfile.BadZipfile: Truncated file header"
11:19:38Araqmy emcc "binary" contains this
11:19:42Araq#!/usr/bin/env python
11:19:42Araq# This script should work in python 2 *or* 3. It loads emcc.py, which needs python 2.
11:19:58Araqand /usr/bin/env python gives me python 2.7.7
11:20:25Araqso why would it produce a SyntaxError? if expressions in Python are surely in 2.7.7
11:21:06FromGitter<Varriount> Araq: Worked on the second try for me. :/
11:21:36FromGitter<Varriount> Araq: I suspect it's somehow using the system version of python, which I believe is 2.7
11:21:41FromGitter<Varriount> er, 2.6
11:31:48euantorInterestingly, `nim doc` works, but not `nim doc2`
11:32:47FromGitter<Varriount> Araq: Is your python installed from brew?
11:52:43Araqit used to be but I switched to pyenv
11:54:43Araqbut now I noticed it's VS's terminal, it didn't pick up my other python
11:54:51Araqso ... now the error is:
11:55:17AraqWARNING:root:LLVM version appears incorrect (seeing "7.0", expected "3.9")
11:55:17AraqCRITICAL:root:fastcomp in use, but LLVM has not been built with the JavaScript backend as a target, llc reports:
11:56:52zachcarterhttp://162.243.192.65/
12:01:38FromGitter<stisa> zachcarter: `failed to load resoure`
12:03:13zachcarterdamn
12:03:50FromGitter<stisa> it's trying to load from `http://162.243.192.65/node_modules/spectre.css/docs/dist/spectre.min.css`
12:08:49zachcartershould be working now
12:09:20zachcarterugh nevermind
12:09:22zachcarterneed to fix one more thing
12:10:24zachcarteryay works now I think
12:10:28zachcarterstisa: mind verifying?
12:10:45FromGitter<stisa> POST http://localhost:5000/compile net::ERR_CONNECTION_REFUSED at app.js:1892
12:11:15FromGitter<stisa> nevermind, works now
12:11:18zachcartersweet :D
12:11:22Tiberiumworks for me
12:11:27Tiberiumehm
12:11:31Tiberium"missing semicolon" ?
12:11:40zachcarteroh
12:11:41FromGitter<stisa> Nice! But damn, almost 500kb for Ace
12:11:41zachcarterwell
12:11:50zachcarterace has no nim language extension
12:11:50Tiberiummmm, *dreaming about integrating nimsuggest into this playground thing*
12:12:12zachcarterI’m not sure there is currently a js editor out there with a nim language extension
12:13:16FromGitter<stisa> codemirror with this : https://github.com/zah/nim.lt/blob/master/codemirror/nimrod.js but I think it's GPL3
12:13:37Tiberiumalso maybe in the future you could somehow not just send compile log to browser when compilation ends, but stream it?
12:13:43Tiberiumor make some progress bar
12:13:46FromGitter<BontaVlad> @ace size, it's not minified, the size could be lowered
12:13:56zachcarterthat will be difficult Tiberium
12:14:06Tiberiumzachcarter, then just some spinning circle :)
12:14:13zachcarterI can do that
12:14:14Tiberiumat least that we know that request have been sent
12:14:20zachcarterI have a lot of things to add
12:15:28FromGitter<BontaVlad> zachcarter, do you have a public repo for the code?
12:15:34zachcarterI do
12:15:44FromGitter<BontaVlad> care to share the url? :)
12:15:50FromGitter<stisa> zachcarter : why are compiler log and program result `<textarea>` ? I'm not good at web dev, but I think a <pre> block is enough?
12:16:03zachcarterhttps://github.com/zacharycarter/nim-playground & https://github.com/zacharycarter/nim-playground-frontend
12:16:12SentreenIs there an easy way to pas a pointer to a var as a var argument? So if I have `proc foo(bar: var Type)`, can I pass a `ptr Type` to it somehow?
12:16:17FromGitter<BontaVlad> many thanks
12:16:20zachcarternp!
12:16:28zachcarterstisa: hrm potentially? I’m not sure
12:16:32*krux02 joined #nim
12:16:39AraqSentreen: deref it, foo(x[])
12:16:49zachcarterI wanted them to be resizable so I went with textarea
12:18:17zachcarterhttps://ace.c9.io/tool/mode_creator.html
12:18:17SentreenCan't believe I didn't think of that.
12:18:22zachcarterlooks like we can create a nim mode
12:18:52FromGitter<stisa> zachcarter oh alright then, Edge doesn't show the resize corner thing ( handler? ) , but chrome does
12:18:53SentreenWill the compiler be smart enough to pass the original pointer as the "hidden" var pointer?
12:19:36zachcarterstisa: might be an issue with spectre css not sure
12:20:55zachcarterI have no idea how to even start writing a nim mode for ace
12:21:38zachcarterI took a look at the python mode and it’s quite complex
12:23:10AraqSentreen: yeah but even if not, &*x is optimized away by every C compiler and the standard mandates it even. C's preprocessor wouldn't work as well otherwise
12:23:41SentreenOkay, thanks again :)
12:25:25Araqok, i give up, back to windows
12:26:00FromGitter<Varriount> Araq: Don't feel too bad. I got the webpage to load in Chrome, but the web console kept giving me a buffer error
12:27:36FromGitter<Varriount> Araq: Quick question - what effect does 'closureScope' have on backend code?
12:28:06FromGitter<Varriount> The template appears to create a single-use closure procedure , which seems rather inefficient.
12:35:49Araqyeah, don't use it for performance
12:39:03*adeohluwa quit (Quit: Connection closed for inactivity)
12:39:10FromGitter<Varriount> Araq: What would it take for a more efficient technique?
12:39:33Araqdunno. escape analysis and closure removal?
12:40:13FromGitter<Varriount> I mean, is closureScope something that could be done directly by the compiler, rather than through a template?
12:41:51Araqthe compiler used to do it, it was a pita
12:42:52FromGitter<Varriount> zachcarter: If you want, feel free to use the tmLanguage file from NimLime
12:43:03zachcarterthank you
12:43:16zachcarterI’ll most likely take you up on that
12:43:28*aedigix quit (Remote host closed the connection)
12:43:33FromGitter<Varriount> It's not perfect, but I think it does an adequate job.
12:43:52*aedigix joined #nim
12:50:16couven92Uhm... trying to compile nim on my android... I managed to compile from csources, but now koch is struggling to boot nim. I follow https://github.com/nim-lang/Nim/pull/5646 I get an error from gcc: `nimcache/r_linux_arm/stdlib_os.c:1824:18: error: no member named 'st_mtim' in 'struct stat'`
12:53:23*zachcarter quit (Quit: zachcarter)
12:57:15Tiberiumcouven92, yes
12:57:21Tiberiumyou need some workarounds
12:57:27TiberiumI'll post a link
12:57:53Tiberiumhttps://github.com/nim-lang/Nim/pull/5646#issuecomment-290931873
12:58:10Tiberiumstarting from "my steps"
12:58:31Tiberiumalso for nimble you want to patch getTempDir()
12:58:46Tiberiumto return "/data/data/com.termux/files/tmp" instead of "/tmp/"
13:00:39*TheLemonMan joined #nim
13:07:01couven92ok, I got it to work now... Nim is surprisingly fast on my 3 years old HTC One m8!! :D
13:12:24*McSoFake quit (Ping timeout: 240 seconds)
13:15:05*bjz_ joined #nim
13:16:29*bjz quit (Ping timeout: 260 seconds)
13:36:42*zachcarter joined #nim
13:43:55couven92Uhm... with Nim I just compiled karax and ran it on my Android! Weeeee! :D
13:44:17*vivus joined #nim
13:44:41FromGitter<Varriount> couven92: You ran the compiler on your phone?
13:44:46couven92yup
13:45:06FromGitter<Varriount> O_o
13:45:43couven92Works beautifully, but I haven't gotten round to compiling nimble yet... Have go to lunch now first :)
13:46:02TiberiumVarriount: see https://github.com/nim-lang/Nim/pull/5646#issuecomment-290931873
13:46:03FromGitter<Varriount> Note to self: couven92 is dangerous
13:46:09TiberiumI did it three days ago
13:46:14Tiberiumstisa did it even earlier
13:47:09Tiberiumcouven92, just patch lib/pure/ospaths.nim - replace "/tmp/" with "/data/data/com.termux/files/tmp/"
13:47:22Tiberiumand nimble should compile now
13:47:27Tiberium(and run)
13:47:31FromGitter<Varriount> Why would you all ever want to run the compiler on a smartphone?
13:47:49Tiberiumbecause why not?
13:48:05Tiberiumfor example, I can run my app on a phone
13:48:07Tiberiumisn't it good?
13:48:10Tiberium(it's console app)
13:48:20vivusDoes Nim support linux environment variables?
13:48:25Tiberiumyes
13:48:30TiberiumAFAIK
13:48:58Tiberiumvivus, https://nim-lang.org/docs/os.html#getEnv,string probably
13:51:47FromGitter<couven92> Well, PMunch and I were pretty impressed that it worked.
13:52:36TiberiumI think we can even contribute it to termux-packages
13:52:49Tiberiumso new users would just type "apt install nim"
13:52:54Tiberiumah no
13:52:54FromGitter<couven92> I'm considering to write up a PR for niminst that adds the necessary changes to build.sh to bootstrap nim
13:53:32vivusif you're writing something to install nim, it should be 10x easier than what is currently available
13:53:59vivusTiberium: a PPA for nim and nimble might be really nice
13:54:07Tiberiumvivus, there's no PPA in termux
13:54:12Tiberiumor I'm mistaken
13:54:34Tiberiumah yes, we can create custom PPA for termux only
13:54:36FromGitter<couven92> Because all in all there weren't many changes I had to do. Most of the time went to typing on the touch keyboard of my phone
13:54:43Tiberiumyeah
13:54:58Tiberiumyou can also use adb if you have root (to launch termux shell)
13:55:02Tiberiumfrom computer via usb
13:55:26FromGitter<couven92> Yeah I know
13:55:35TiberiumI think we can create a simple .sh script for now :)
13:56:43vivuswell at least 0.16.0 has remained stable . is there a release-schedule?
13:56:56FromGitter<couven92> No, I'll change niminst build.sh then we're good to go
13:57:24Tiberiumcouven92: but we also need to change nim files
13:57:27Tiberiumwill build.sh handle this?
13:59:35FromGitter<couven92> Yes, and another PR to enable Android support in two places in the nim source
13:59:45FromGitter<couven92> That's it
14:02:47*TheLemonMan quit (Quit: "It's now safe to turn off your computer.")
14:04:36Tiberiumcouven92: I also thought about that, but later I realized that termux isn't a big platform, so I thought that PR would not be accepted :)
14:07:06*devted joined #nim
14:12:36*couven92 quit (Quit: Client disconnecting)
14:17:05*chemist69 quit (Ping timeout: 260 seconds)
14:20:54*chemist69 joined #nim
14:23:29Tiberiumby the way, what style is preferred? "var a = 1; var b= 2" or one "var" block like "var\n a=1\nb=2" ?
14:24:45PMunchTiberium, I prefer, and I think it's more ideomatic, to put it in one block.
14:25:31Tiberiumand also, can I contribute some small changes to base64 (putting variables into one "var" or "let" block, and changing "var" to "let" where variables aren't changing)?
14:25:39PMunchBut I have split stuff into two blocks when they are related to each other. So if I'm using two frameworks and need to set some variables for one, and some for another then I might split those up into two different blocks.
14:26:01Tiberiumbecause my changes are small
14:26:15PMunchTiberium, I don't see why not. Worst case scenario they don't get pulled :P
14:26:39*BennyElg quit (Remote host closed the connection)
14:27:05*BennyElg joined #nim
14:30:01*couven92 joined #nim
14:30:14TiberiumPMunch, https://github.com/nim-lang/Nim/pull/5755/commits/5ee75a66aca0205bb2ab72704473a94cd41e575a
14:31:36*BennyElg quit (Ping timeout: 260 seconds)
14:32:47PMunch79 var -> let?
14:33:05PMunchSame for 89
14:33:10PMunchlines that is
14:33:15TiberiumPMunch, oh :)
14:33:49PMunchBut otherwise it looks gode
14:33:52PMunchgood*
14:34:21PMunchBut you removed the {.immediate.} pragma on line 47, was that on purpose?
14:34:44Tiberiumoh sorry
14:36:04TiberiumPMunch, that's why I tried to remove immediate: "base64.nim(47, 75) Warning: use 'untyped' parameters instead; immediate is deprecated"
14:36:18PMunchAh right
14:36:28PMunchI didn't know what it was, just saw that you removed it :P
14:36:42PMunchBut then it's fine
14:36:59Tiberiumlet's wait for tests :D
14:37:56Tiberium(sorry I will not spam commits like that in future)
14:40:03Tiberiumoh I think we gonna have long time with tests
14:41:02*PMunch quit (Quit: leaving)
14:49:44*BennyElg joined #nim
14:59:14*jivank[m] quit (K-Lined)
14:59:15*ehmry[m] quit (K-Lined)
14:59:15*stisa quit (K-Lined)
14:59:16*Jipok[m] quit (K-Lined)
14:59:16*hohlerde quit (K-Lined)
14:59:17*erwana[m] quit (K-Lined)
14:59:17*TheManiac quit (K-Lined)
14:59:17*M-Quora quit (K-Lined)
14:59:17*mindB quit (K-Lined)
14:59:17*Guest73656[m] quit (K-Lined)
14:59:17*MrAxilus[m] quit (K-Lined)
14:59:18*dyce[m] quit (K-Lined)
15:01:36*BennyElg quit (Remote host closed the connection)
15:05:03*samuell joined #nim
15:06:06*PMunch joined #nim
15:10:53*dyce[m] joined #nim
15:16:08zachcarteralright all requests to the nim playground should now get their own tmp directory
15:16:09*MrAxilus[m] joined #nim
15:16:09*M-Quora joined #nim
15:16:09*mindB joined #nim
15:16:10*ehmry[m] joined #nim
15:16:15*stisa joined #nim
15:16:15zachcarterstill have to add reqest caching
15:16:15*Guest73656[m] joined #nim
15:16:15*TheManiac joined #nim
15:16:16*jivank[m] joined #nim
15:16:16*hohlerde joined #nim
15:16:16*erwana[m] joined #nim
15:16:17*Jipok[m] joined #nim
15:16:57zachcarterI may have just broken things too
15:19:31*Snircle quit (Quit: Textual IRC Client: www.textualapp.com)
15:22:59*gokr quit (Quit: Leaving.)
15:24:28*sz0 quit (Quit: Connection closed for inactivity)
15:28:50*couven92 quit (Ping timeout: 240 seconds)
15:29:19*couven92 joined #nim
15:31:55*Trustable joined #nim
15:33:41*rauss joined #nim
15:34:07*nsf quit (Quit: WeeChat 1.7)
15:35:55Tiberiumzachcarter, 502 again :)
15:36:08zachcarteryeah :/ working on the server again
15:36:27Tiberiumzachcarter, also it gives "XMLHttpRequest cannot load http://162.243.192.65:3000/compile. No 'Access-Control-Allow-Origin' header is present on the requested resource. Origin 'http://162.243.192.65' is therefore not allowed access. The response had HTTP status code 502."
15:36:47zachcarterI don’t even think the service is running atm
15:37:40*bjz_ quit (Quit: My MacBook Pro has gone to sleep. ZZZzzz…)
15:38:11vivuswhen it is said "nim compiles to C" does it actually mean nim is being transpiled to C code and then compiled?
15:39:10Tiberiumvivus, don't use word "transpile" please
15:39:42vivusTiberium: that is why I am asking what is actually happening
15:40:52Tiberiumvivus, it compiles to C and then C compiler compiles C code :)
15:41:13Tiberiumif Nim would be something very similar to C, it would be possible to say "transpiles to C"
15:41:18Tiberiumbut Nim is very different
15:41:43Tiberiumhttps://www.typescriptlang.org/ even typescript "compiles"
15:41:48Tiberiumto JS
15:42:19zachcarterTiberium: should be working again
15:42:26vivusthanks for clearing that up. I believe I can also compile to LLVM?
15:42:38Tiberiumzachcarter, yep
15:42:50zachcartersweet!
15:42:51Tiberiumvivus, afaik not really, but someone is working on it
15:43:45FromGitter<stisa> vivus https://github.com/arnetheduck/nlvm if you are interested, that's the llvm backend
15:43:57*Jesin joined #nim
15:44:13vivuswhat advantages would compiling direct to LLVM have over compiling to C?
15:45:27Tiberiumwat
15:45:28Tiberiumhttps://github.com/arnetheduck/j2c
15:45:43Tiberiumah, java
15:45:45TiberiumI thought JS :D
15:46:19dom96Actually for typescript I would say that "transpiling" is the correct term
15:46:37Tiberiumyep
15:46:43Tiberiumbecause it understands JS syntax fully
15:46:47Tiberiumand it's built on top of it
15:46:57dom96Take a look at the wikipedia page for "transpiler"
15:47:24Tiberiumso nim compiler is transpiler?
15:47:33Tiberiumah sorry :)
15:47:40Tiberiumwhile a traditional compiler translates from a higher level programming language to a lower level programming language
15:48:05dom96indeed
15:48:16dom96JS and Typescript are both the same level of abstraction
15:48:26dom96Well, I guess you could say that they aren't
15:48:48Araq"transpiling" is still a pointless word
15:48:48dom96*are both at
15:49:22Araqit's called "translator" or "compiler" or "assembler", there is no need for a "transcompiler", that's just bullshit
15:49:44TiberiumAraq, I think this word became popular from JS world?
15:51:05TiberiumAlso, when someone say "Nim git repository has almost 1 thousand issues!", point him to https://github.com/Microsoft/TypeScript/issues :D
15:51:41Araqmost of the classical optimizations on a decent IR could be performed by TypeScript targetting JS.
15:52:17Araqmost of the high level vs low level distinctions are expressible in any programming language
15:53:04Araqloop invariant code motion, common subexpression elimination, loop hoisting etc etc can be done by a C to C "transpiler".
15:54:29vivuswhat advantages would compiling direct to LLVM have over compiling to C?
15:54:57Araqthe only real difference is that most programming languages (including C) are really bad backend target languages
15:56:35Araqvivus: there is no consensus. arguably, for Nim: better debugging support, faster exception handling and a possibly faster GC
15:57:29vivuswhich would give those options? LLVM or C?
15:57:37TiberiumI think LLVM
15:59:13subsetparkDoes the stdlib have a proc for making DNS requests?
16:01:45dom96subsetpark: very low level, but getAddrInfo exists
16:01:57dom96PRs to add a higher-level version to the next module are welcome
16:02:04dom96*net module
16:02:29dom96zachcarter: how's the site going?
16:02:59subsetparkdom96: Cool, sounds like a nice target
16:10:00*gokr joined #nim
16:10:10couven92exit
16:10:16couven92oh... fu***
16:10:22Tiberium:D, wrong window?
16:10:29couven92Tiberium, yup
16:17:09*BennyElg joined #nim
16:21:44*chemist69 quit (Ping timeout: 240 seconds)
16:22:08*chemist69 joined #nim
16:24:46*Jesin quit (Ping timeout: 245 seconds)
16:26:19Tiberiumcan we get a directory where source file is located?
16:26:36Tiberiumfor example I have "test.nim" in "/home/tiber/folder"
16:26:41Tiberiumand I want to get "/home/tiber/folder"
16:28:15*brson joined #nim
16:32:25*gokr quit (Ping timeout: 255 seconds)
16:39:41*couven92 quit (Quit: Client disconnecting)
16:42:52FromGitter<stisa> Tiberium : look at https://nim-lang.org/docs/system.html#instantiationInfo,
16:43:53*xet7 quit (Quit: Leaving)
16:43:54*couven92 joined #nim
16:48:30SentreenIs there any particular reason why nim handles circularity so poorly? (i.e. why recursive module imports don't work and why procs need to be forward declared)? It's something that keeps on coming back to bite me in the ass
16:49:23*yglukhov quit (Remote host closed the connection)
16:51:35SentreenIt seriously baffles me that this sort of thing is not on the 1.0 roadmap, so I assume there's a reason for it
16:52:03*yglukhov joined #nim
16:56:42*yglukhov quit (Ping timeout: 264 seconds)
17:04:15*samuell quit (Quit: Hejdå)
17:04:21*xet7 joined #nim
17:05:28FromGitter<Varriount> Sentreen: Compiler complexity.
17:05:31zachcarterdom96: good but me creating a nim mode for ace editor is not
17:05:42zachcarterdom96: http://162.243.192.65/
17:06:34dom96cool
17:06:40dom96You need some spinners :)
17:06:46zachcarterhaha Tiberium mentioned that
17:06:49zachcarteri will add some this evening
17:07:18zachcarterI think I’m going to leave creating a nim mode for ace up to someone else i don’t think I’m going to be able to do it
17:07:29euantorWas looking at the streams module earlier, and noticed something that I miss from the .net world - a `copyTo` procedure to copy contents from one stream to another
17:07:31zachcarterseems bizzrarely complex
17:07:46FromGitter<Varriount> Sentreen: I can't speak for circular imports, but requiring that procedures be forward declared simplifies quite a bit of the compiler logic.
17:08:06*GustavoLapasta joined #nim
17:08:11euantorAnd of course on systems that support it that proc could use sendfile when working with an IO stream
17:08:51euantorWhich could then be used with sending data from a file to a HTTP client with a http server or reading from a TCP client into a file without doing it manually
17:09:07dom96zachcarter: yeah, no need
17:09:12euantorOf course, that would require sockets/http client to use streams rather than strings but that's something I would like to see for definite
17:09:14dom96just enable Python highlighting for now
17:09:19zachcarterdom96: alright
17:09:33dom96although IIRC ace supports textmate syntax highlighting
17:09:39dom96and there is a Nim one somewhere
17:09:55zachcarteroh it does? let me find that then
17:11:06*Ven joined #nim
17:11:30*Ven is now known as Guest25397
17:11:45*PMunch quit (Quit: leaving)
17:12:17*Guest25397 quit (Client Quit)
17:14:06SentreenThat's a shame, but it makes sense :).
17:15:28dom96Sentreen: Removing forward declarations is on our roadmap
17:15:28FromGitter<Varriount> Sentreen: Though, generally circular imports are frowned upon, even in languages that support them fairly weel.
17:15:31FromGitter<Varriount> *well
17:18:05dom96We need to publish an updated roadmap :)
17:18:10euantorzachcarter: I'm getting the following error on the playground: ace.edit can't find div #editor
17:18:19SentreenAh, I didn't know it was outdated :).
17:18:51SentreenI know that it's generally a design smell, but sometimes the only alternative is throwing everything in one huge file, which I personally hate :).
17:21:56dom96I've been able to get away with a 'types' module in (IIRC) all cases.
17:22:04dom96Maybe you could do the same?
17:23:37SentreenI just switched to something like that, just dealing with a few procedure calls that cross module boundaries.
17:24:34SentreenI just asked because it felt like I was struggling with the language, not with the actual problems, which is something I generally don't have when I'm writing nim :)
17:27:41zachcartereuantor: hrm
17:29:01zachcarterI got the error too but it seems sporadic
17:29:03dom96Sentreen: that's fair, there likely is a way to improve this but I am not sure how. Suggestions welcome :)
17:30:12FromGitter<Varriount> dom96: I think Nim's module model could be improved a bit.
17:35:27SentreenI think forward declarations would already help a great deal. Some way to have types depend on each other without generics (maybe by forward declaring types?) would also help a great deal. But I don't know enough about the compiler infrastructure to know the implications of something like that.
17:36:35FromGitter<Varriount> Sentreen: Types can be circular if they are in the same type block.
17:37:15*yglukhov joined #nim
17:37:49TiberiumIs there an opposite to https://nim-lang.org/docs/xmltree.html#escape,string in stdlib?
17:38:03SentreenSorry, I meant making that possible across module boundaries. But I don't think Araq was a fan of even doing it outside of a single type block.
17:38:47FromGitter<Varriount> I'm not a fan either. If two types depend on each other, they should be close together (in the same module)
17:40:56FromGitter<Varriount> @zacharycarter Present for you: https://github.com/amoffat/dockerbomb
17:41:07SentreenThat works for small type compositions, but it quickly becomes a mess if you have a few large (as in a lot of functionality) types that depend on each other. When that happens making a separate 'type' file (as dom said) seems to be the only real solution.
17:41:16euantorzachcarter: I think it's because the code is running before the DOM is loaded. Usually you'd use an event for document ready to invoke any code relying on the DOM
17:41:26*yglukhov quit (Ping timeout: 245 seconds)
17:41:33SentreenThis might just be really hating huge files though.
17:41:37zachcartereuantor I just tried pushing an update to make that happen
17:41:41zachcarterbut it doesn’t look like it worked
17:42:19Araqif you have a large type graph full of cycles you have a mess really ;-)
17:42:39FromGitter<Varriount> Araq: You mean, like the compiler?
17:42:42FromGitter<Varriount> :P
17:43:27FromGitter<Varriount> I'm kidding. The compiler has a complex type graph because of it's nature.
17:44:06SentreenI'd argue that the nim compiler is not the only system with a complex type graph, even though I agree that quickly becomes a mess :P
17:47:39zachcarterAraq: is there any way to do things post rendering the vdom with karax?
17:48:04zachcarterI tried this but it doesn’t seem to be working : https://github.com/zacharycarter/nim-playground-frontend/blob/master/src/app.nim#L42-L75
17:50:41*TheLemonMan joined #nim
17:56:22*adeohluwa joined #nim
17:58:53*vendethiel- quit (Ping timeout: 240 seconds)
18:00:11*GustavoLapasta quit (Quit: Leaving)
18:02:51*Ven joined #nim
18:03:15*Ven is now known as Guest3812
18:04:26*couven92 quit (Quit: Client disconnecting)
18:07:32Araqzachcarter: why not?
18:07:58zachcarternot sure - keeep getting missing resource errors
18:08:14AraqloadScript "src/ace.js" # well I dunno what ace.js does
18:09:02zachcartersorry the error is not missing resource the error is that an element is not found in the DOM
18:09:07zachcarterhere’s the content of ace.js
18:09:39zachcarterhttps://github.com/zacharycarter/nim-playground-frontend/blob/master/src/ace.js
18:09:50zachcarterline 3 is the culprit
18:10:04zachcarterthe error is that an element with the ID editor does not exist
18:10:21zachcarterso it would seem to me this script is being executed before the DOM is actually fully loaded
18:11:27Tiberiumwhy I cant create string table as a const?
18:15:41demi-tables aren't defined structures at runtime?
18:16:38demi-most languages i know of that allow you to `const` a non-primative type apply a hack to make it be interpreted as a const but has to be created at runtime, not compile-time
18:16:50demi-er
18:17:04demi-tables aren't defined structures at compile-time, is what i meant to ask
18:18:06dom96demi-: that's more of a statement than a question :P
18:18:31demi-it is an educated guess as to why you cannot const a table
18:18:35demi-but i don't know for sure
18:18:47Araqwell I can use a const table and it's in the test suite
18:20:25Tiberiumyep, found it
18:21:43*BennyElg quit (Remote host closed the connection)
18:22:01*BennyElg joined #nim
18:29:25*yglukhov joined #nim
18:32:34*Jesin joined #nim
18:34:46*Guest3812 quit (Ping timeout: 245 seconds)
18:34:52*nsf joined #nim
18:35:05Tiberiumhttps://github.com/nim-lang/Nim/pull/5755 yay no test errors
18:35:38*gokr joined #nim
18:37:25*brson quit (Quit: leaving)
18:37:31*Ven_ joined #nim
18:40:23*zachcarter quit (Ping timeout: 240 seconds)
18:45:59krux02I found a new bug in the Nim language, just created the issue
18:46:12krux02yay
18:47:26dom96Tiberium: cool, merged
18:47:48TheLemonManhow comes everyone's interested in matrices?
18:48:29dom96Tiberium: note that using multiple lets/vars is generally ok.
18:48:57Tiberiumdom96, ok, but someone said that it's more idiomatic to use blocks
18:51:00krux02TheLemonMan: you style of words let me think that you are not interested in matrices
18:52:17krux02matrices are cool, because they describe linear mappins in just a bunch of numbers plus a lot of theory about them that allow very defined transformations
18:54:33TheLemonManmore like because I've been sifting trough the bug tracker and there are a lot of bugs involving static[T] and people trying to model matrices
19:09:07*Arrrr quit (Ping timeout: 258 seconds)
19:19:42*zachcarter joined #nim
19:27:48*rauss quit (Ping timeout: 246 seconds)
19:30:25*rauss joined #nim
19:32:17*rokups quit (Quit: Connection closed for inactivity)
19:34:02*brson joined #nim
19:34:23Tiberiumwow https://github.com/nim-lang/Nim/issues/5757
19:36:01*yglukhov quit (Remote host closed the connection)
19:36:42Tiberiumhmm wait
19:37:10Tiberiumit compiles on nim playground, but not on my PC
19:39:31*yglukhov joined #nim
19:39:39*Trustable quit (Remote host closed the connection)
19:39:42*yglukhov quit (Remote host closed the connection)
19:40:59TheLemonManmore like because I've been sifting trough the bug tracker and there are a lot of bugs involving static[T] and people trying to model matrices
19:41:05*TheLemonMan left #nim (#nim)
19:41:31dom96Tiberium: known regression (maybe)
19:52:46*yglukhov joined #nim
20:03:36*yglukhov quit (Remote host closed the connection)
20:06:54*yglukhov joined #nim
20:09:03*adeohluwa quit (Quit: Connection closed for inactivity)
20:17:43*couven92 joined #nim
20:23:42Tiberiumdom96, can I somehow disable error throw at 404 error code? using async http client
20:23:45Tiberiumor set up a callback
20:24:22dom96use request
20:24:58*zachcarter quit (Quit: zachcarter)
20:42:24*Tiberium quit (Remote host closed the connection)
20:46:48*Vladar quit (Quit: Leaving)
20:52:02*Ven_ quit (Ping timeout: 252 seconds)
20:53:28*Ven joined #nim
20:53:47*couven92 quit (Quit: Client Disconnecting)
20:53:52*Ven is now known as Guest73192
20:57:04FromGitter<ephja> the website is looking good
20:59:58*rauss quit (Quit: WeeChat 1.7.1)
21:01:58*bjz joined #nim
21:03:01*Matthias247 joined #nim
21:05:55*PMunch joined #nim
21:14:24*libman joined #nim
21:18:40*Guest73192 quit (Quit: My MacBook has gone to sleep. ZZZzzz…)
21:30:20*gokr quit (Quit: Leaving.)
21:58:30*devted quit (Quit: Sleeping.)
22:04:48*vlad1777d_ joined #nim
22:04:49*zetashift joined #nim
22:05:04*scriptum quit (Ping timeout: 240 seconds)
22:07:07*vlad1777d quit (Ping timeout: 268 seconds)
22:07:56*gangstacat quit (Ping timeout: 252 seconds)
22:08:04*scriptum joined #nim
22:09:46*BennyElg quit (Read error: Connection reset by peer)
22:10:33*BennyElg joined #nim
22:11:41*Matthias247 quit (Read error: Connection reset by peer)
22:13:28*gangstacat joined #nim
22:14:44*nsf quit (Quit: WeeChat 1.7)
22:22:54*themagician quit ()
22:23:08*yglukhov quit (Read error: Connection reset by peer)
22:23:39*yglukhov joined #nim
22:23:52*zetashift left #nim ("Leaving")
22:34:56*bjz quit (Quit: My MacBook Pro has gone to sleep. ZZZzzz…)
22:40:10*PMunch quit (Quit: leaving)
22:44:10*yglukhov quit (Remote host closed the connection)
22:44:42*yglukhov joined #nim
22:46:06FromGitter<TiberiumPY> LOL nim has phone back end
22:46:32FromGitter<TiberiumPY> *php
22:49:15*yglukhov quit (Ping timeout: 260 seconds)
22:58:12FromGitter<TiberiumPY> And nim ffi is again so easy: default function to read import from console in PHP is readline so you just do "proc readline(prompt: cstring): cstring {.importc.}" and it just works
23:00:06FromGitter<TiberiumPY> s/import/input
23:08:50*zachcarter joined #nim
23:12:10*smt_ joined #nim
23:12:39*devted joined #nim
23:15:10*smt quit (Ping timeout: 255 seconds)
23:15:40*zachcarter quit (Ping timeout: 260 seconds)
23:16:59*zachcarter joined #nim
23:21:33*sz0 joined #nim
23:43:17*vopi181[m] joined #nim
23:55:24zachcarterman so weird the nim playground didn’t seem to be having this issue loading JS earlier today
23:55:28zachcarternot sure what I broke :/